In ChandlerLutz/starpolishr: polish stargazer (latex) files

knitr::opts_chunk$set(echo = TRUE)

First note that the first argument in all starpolishr functions is a stargazer table, allowing for easy compatibility with the magrittr %>%. Also, all of the functions in the starpolishr package begin with star_ for easy tab completion in emacs or rstudio.

Basic Stargazer Tables

To start, let's run a couple of regressions. These will be miles per gallon, mpg, on hp and cyl from the mtcars dataset. We'll also produce a basic stargazer table.

library(stargazer); library(starpolishr); library(magrittr)
data(mtcars)
mod.mtcars.1 <- lm(mpg ~ hp, mtcars)
mod.mtcars.2 <- lm(mpg ~ hp + cyl, mtcars)
star.out.1 <- stargazer(mod.mtcars.1, mod.mtcars.2,
                        title = "Motor Car Regressions",
                        type = "latex",
                        #For exact latex positioning using the latex 'float' package
                        table.placement = "H",
                        keep.stat = "n",
                        header = FALSE)

Obviously, a potential confound is the vehicle weight, wt. For the purposes of this vignette, let's say we want the regressions that employ the wt variable in a separate table:

#Second set of models with weight as a regressor
mod.mtcars.3 <- lm(mpg ~ hp + wt, mtcars)
mod.mtcars.4 <- lm(mpg ~ hp + cyl + wt, mtcars)
star.out.2 <- stargazer(mod.mtcars.3, mod.mtcars.4,
                        #For exact latex positioning using the latex 'float' package
                        table.placement = "H",
                        keep.stat = c("n", "rsq"),
                        header = FALSE)

Updating Regression Variable Names

We can improve a number of things about these tables. First, the left-hand and right-hand side variables are a bit cryptic. Let's clean this up using starpolishr star_lhs_names() and star_rhs_names() functions, which use regular expressions to replace variable names. The advantage of these functions is that they allow the variable names to span more than one line. As we want to apply these functions to both tables, let's create a function to minimize duplicative code.

clean_var_names <- function(table) {
    table %>%
        #Update the RHS names
        star_rhs_names(pattern = c("hp", "cyl", "wt"),
                       line1 = c("Hoarsepower", "Number of", "Vehicle"),
                       line2 = c("", "Engine Cylinders", "Weight")) %>%
        star_lhs_names(pattern = c("mpg"),
                       line1 = "Miles per",
                       line2 = "Gallon")
}

Now let's apply the function to the two stargazer tables and print

star.out.1 <- clean_var_names(star.out.1)
star.out.2 <- clean_var_names(star.out.2)

cat(star.out.1)

cat(star.out.2)

\clearpage

Create a table with 2 panels and add custom notes

Now let's combine the tables into a two panel single table using the star_panel() function and add custom table notes using the star_notes_tex() function. Finally, we'll print the output

star_panel(star.out.1, star.out.2,
           panel.names = c("Reg Without Weight", "Reg With Weight"),
           panel.label.fontface = "bold" #For bold panel names
           ) %>%
    #Add table notes
    star_notes_tex(note.type = "caption", #Use the latex 'caption' package for notes
                   note = "Standard errors are in parentheses") %>%
    cat